promptslab/awesome-prompt-engineering

Awesome Prompt Engineering is a curated collection of links, papers, tools, courses, and communities related to prompt engineering: the practice of writing and refining the text instructions you give to AI language models to get better results. It is a reference document, not a piece of software you run. Prompt engineering matters because the same AI model can produce very different outputs depending on how a question or task is phrased. Researchers and practitioners have developed systematic techniques for structuring these instructions: for example, asking a model to think through a problem step by step before answering (called chain-of-thought prompting), or giving it a few worked examples before the actual question (called few-shot prompting). The list gathers papers that study and compare these techniques, including several large surveys covering dozens of approaches. Beyond techniques, the repository links to practical tools: platforms for testing and managing prompts across different AI models, evaluation tools that measure how well a prompt works, agent frameworks for building systems where multiple AI calls happen in sequence, and tools for identifying security risks in prompts (such as prompt injection, where malicious input tries to override the original instructions). Other sections cover the AI models themselves, the APIs used to access them, benchmark datasets for measuring model performance, online courses ranging from beginner to advanced, video tutorials, and active community spaces like Discord servers where practitioners share findings. The list was last updated in February 2026 and is structured with a suggested learning path for newcomers: start with a free short course, read a comprehensive guide, check the official documentation for major AI providers, then dig into the research papers. Contributions from the community are accepted via pull request. The full README is longer than what was shown.